The Relation Between Metabolic Syndrome and Its Components with Breast Density in Postmenopausal Women

Abstract

Background:

Metabolic syndrome (MetS) could lead to an increase in fatty tissue that could be seen as a radiolucent image depicting breast density (BD) by a mammogram. We aimed to investigate the association between MetS and its separate components with BD among naturally postmenopausal women.

Materials and Methods:

Data of 494 postmenopausal patients who were admitted to our outpatient clinic between December 2012 and July 2015 were retrospectively reviewed. A total of 279 patients were in the without MetS group and 215 patients were in the with MetS group. Average BD percentage of the left and right breasts were measured. Basic characteristics, laboratory, and mammography results between the without MetS and the with MetS groups were compared.

Results:

The mean age of the patients was 53.20 ± 6.67 years in the without MetS group and 55.41 ± 6.56 years in the with MetS group. There were 219 (78.5%) patients in the without MetS group and 187 (86.9%) patients in the with MetS group with lower BD. The without MetS group had significantly higher BD scores than those patients in the with MetS group (P = 0.02). In correlation analysis, there was a negative correlation between fasting plasma glucose (FPG), systolic and diastolic blood pressures, waist circumference (WC), and BD scores. However, there was a positive correlation between high-density lipoprotein (HDL) and BD score (P = 0.046). In multivariate logistic regression analysis, it is found that lower body mass index (BMI) and parity were significantly associated with higher BD (P = 0.002 and P = 0.001; respectively).

Conclusion:

The lower BMI and parity may be associated with higher BD in postmenopausal women. In addition, higher HDL and lower FPG, blood pressure, triglyceride, and WC may be correlated with higher BD.

Introduction

Metabolic syndrome (MetS) is a complex disease of various metabolic components including abdominal obesity, high blood pressure, hyperglycemia, and dyslipidemia.¹ The incidence of MetS is ∼35% in the older than 20-year-old population and 41.1% in Turkish women.²

Cardiovascular and metabolic effects of MetS could lead to an increase in fatty tissue that could be seen as a radiolucent image depicting lower breast density (BD) by a mammogram.³ It is also reported that higher BD might reflect a sum of increased exposure to hormones and growth factors.⁴ Moreover, the radiologically confirmed increased BD could be associated with a four- to sixfold increase in breast cancer risk, and it is also reported that a high BD is related to a >60% of breast cancer risk, whereas a low BD is associated with a <10% of breast cancer risk.⁵

Although there are controversial reports with regard to the association between MetS and breast cancer risk,^1,6 in previous studies, postmenopausal women with MetS had a 75% risk of breast cancer compared with women without MetS.⁷

However, little is known about the association between MetS and BD in our region. Therefore, in this study, we aimed to investigate the association between MetS and its individual components with BD among naturally postmenopausal women.

Materials and Methods

This study was performed at our outpatient clinic with retrospective analysis of 494 postmenopausal patients' data between December 2012 and July 2015, after obtaining ethical approval from the hospital's local ethics committee (Protocol Number: 2015-168). An evaluation of the severity of BD among postmenopausal women was made while the women were screened during their annual visit. The exclusion criteria were as follows: use of hormonal therapy, chemotherapy or pelvic radiotherapy owing to any type of cancer, thyroid diseases, Cushing's disease, premature menopause, and surgical menopause after bilateral oophorectomy. The inclusion criteria were as follows: women who did not comprise exclusion criteria, nonhormone therapy users, naturally postmenopausal women aged between 40 and 65 years, and those who read the consent form and accepted to participate in the study.

Postmenopausal status was described as those with at least 12 months of cessation of menstrual periods with any other medical reasons and with a serum follicle-stimulating hormone level of 40 mIU/mL. Taking into account the exclusion criteria, the patients were divided into two groups. There were 279 patients in the without MetS group and 215 patients in the with MetS group. The basic characteristics of the patients [age, parity, gravidity, time since menopause onset, and body mass index (BMI)] were determined. The systolic and diastolic blood pressures were measured from the women's left arms by using a sphygmomanometer (ERKAmeter 3000; ERKA, Bad Tolz, Germany) with a suitable cuff size in the sitting position after a 5-min rest. Blood pressure measurements were taken twice and a mean value was computed. BMI was calculated by dividing the weight (kg) to the square of the height (m²). Waist circumference (WC) was also measured as the minimum length around the umbilicus with a tape placed on a horizontal plane at the point of the iliac crest while in an upright position.

Laboratory measurements

Blood samples were gathered between 08:00 and 10:00 a.m. after a 12-hr fasting period. The sample tubes were centrifuged at 2000 rpm (10 min) and serum samples were kept at 80°C. Fasting plasma glucose (FPG) was studied by the hexokinase method, triglyceride (TG) was analyzed by the TG glycerol phosphate oxidase method, high-density lipoprotein (HDL) was analyzed by the accelerator selective detergent method, total cholesterol (TC) was analyzed by the enzymatic method, and low-density lipoprotein (LDL) was measured by the Friedewald formula on an Abbott Architect auto-analyzer by using their original kits (Abbott Laboratories, Chicago, IL).

Metabolic syndrome

MetS was diagnosed according to the National Cholesterol Education Program Adult Treatment Panel III (NCEP-ATP III) criteria, with the women presenting with three or more of the five risk determinants being diagnosed with MetS. These criteria were as follows: an abdominal obesity, defined as a WC in women >88 cm, serum TG >150 mg/dL (1.7 mmol/L), or a drug treatment for elevated TG, a serum HDL cholesterol of <50 mg/dL (1.3 mmol/L) in women or drug treatment for a low HDL cholesterol, an FPG >100 mg/dL (5.6 mmol/L) or a drug treatment for elevated blood glucose, and a blood pressure >130/85 mmHg or a drug treatment for an elevated blood pressure.⁸

Mammography

All mammograms were acquired with a full-field digital mammography (Hologic Lorad Selenia; Danbury), and synthetic mammograms were generated by using the U.S. Food and Drug Administration-approved “C-View” software module. The “For Presentation” standard dose mammograms and synthetic images were analyzed using a fully automated algorithm. All mammograms were acquired in standard craniocaudal and mediolateral oblique projections with automatic image acquisition parameters. The level of compression was determined by an experienced mammographic technician. Average percent BD was defined as the ratio of fibroglandular tissue to total breast volume for both sides of the breasts and assigned to Breast Imaging Reporting and Data System (BI-RADS) categories of 0–4 resembled by increasing density.⁹ In this study, we classified patients according to their BD levels. The BI-RADS 0–2 cases were considered to be low BD and the BI-RADS 3–4 cases were considered to be high BD in accordance with the previous study.³

Statistical analysis

Data analysis was performed with SPSS (version 20.0; Chicago, IL). All the data were presented as mean ± standard deviation. One sample K–S test was performed to analyze the distribution of the data and logarithmic transformations were used where appropriate. Basic characteristics, laboratory, and mammography results between the without MetS and with MetS groups were compared using Student's t-test for parametric variables, and Mann–Whitney U test (two independent samples) for nonparametric variables. Point biserial correlation tests were used to determine correlations between BD and MetS variables for all the study population. The predictive variables of higher BD were determined by a multivariate linear regression test. A regression model was conducted by means of stepwise procedure and comprised all potential variables. The model was generated from independent variables providing P = 0.10 at bivariate analysis and then the best-fit model was performed without any interacting variables. BD scores were accepted as dependent variables. Independent variables considered were age, parity, gravidity, time since the menopause onset, FPG, HDL, and TG. For all calculations, statistical significance was accepted as P < 0.05.

Results

The mean age of the patients was 53.20 ± 6.67 years in the without MetS group and 55.41 ± 6.56 years in the with MetS group. The mean gravidity and parity was 3.8 ± 2.2 and 2.41 ± 1.35 in the without MetS group and 4.73 ± 2.53 and 3.09 ± 1.76 in the with MetS group, respectively. There were significantly higher numbers of age, gravidity, parity in the with MetS group compared with the without MetS group (P < 0.001, P < 0.001, P < 0.001, respectively). There was no significant difference with regard to time since menopause, TC, and LDL levels between the study groups. All individual MetS parameters, namely FPG, TG, systolic or diastolic blood pressures, HDL, and WC were significantly different for the without MetS compared with the with MetS groups (Table 1). When it comes to lower BD scores, there were 219 (78.5%) patients in the without MetS group and 187 (86.9%) patients in the with MetS group, and there were 60 (21.5%) patients in the without MetS group and 28 (13.1%) patients in the with MetS group with higher BD scores and the without MetS group had significantly higher BD scores than those patients in the with MetS group (P = 0.02) (Table 2). Furthermore, correlation analysis depicted negative relations between FPG, systolic and diastolic blood pressures and WC with BD scores. However, there was positive association between HDL and BD scores (P = 0.046) (Table 3). After adjustments, multivariate regression analysis considering age, gravidity, parity, BMI, and individual MetS parameters, there was a significant effect of parity and BMI on BD scores (P = 0.001, 95% confidence interval [CI] = 0.41–0.78; P = 0.002, 95% CI = 0.83–0.95, respectively) (Table 4).

Table 1.

Comparison of the Basic Characteristics and Laboratory Analysis of the Study Groups

	Without MetS (n = 279)	With MetS (n = 215)	P
Age (years)	53.20 ± 6.66	55.41 ± 6.56	0.000
Gravidity	3.80 ± 2.22	4.73 ± 2.53	0.000
Parity	2.41 ± 1.34	3.09 ± 1.76	0.000
Time since menopause onset (months)	85.21 ± 64.53	98.29 ± 76.12	0.117
TC (mg/dL)	212.96 ± 37.11	217.03 ± 42.11	0.327
LDL (mg/dL)	165.58 ± 55.25	133.44 ± 37	0.929
MetS parameters
FPG (mg/dL)	91.58 ± 14.91	108.73 ± 35.44	<0.001
Blood pressure (mm/Hg)
Systolic	114.81 ± 14.38	127.91 ± 18.35	<0.001
Diastolic	71.86 ± 9.4	79.53 ± 10.11	<0.001
Log-HDL (mg/dL)	17.72 ± 0.01	16.66 ± 0.1	<0.001
Log-TG (mg/dL)	19.85 ± 0.15	22.28 ± 0.17	<0.001
WC (cm)	100.35 ± 13.17	104.76 ± 12.87	<0.001
BMI (kg/m²)	28.15 ± 4.5	30.22 ± 5.6	<0.001

Bold values: P < 0.05.

BMI, body mass index; FPG, fasting plasma glucose; HDL, high-density lipoprotein; LDL, low-density lipoprotein; MetS, metabolic syndrome; TC, total cholesterol; TG, triglyceride; WC, waist circumference.

Table 2.

Comparison of the Severity of Breast Density Between the Study Groups

	Low BD, n (%)	High BD, n (%)	P
Without MetS	219 (78.5)	60 (21.5)	0.02
With MetS	187 (86.9)	28 (13.1)	0.02

Bold values: P < 0.05.

BD, breast density.

Table 3.

Points Biserial Correlation Analysis of Individual Metabolic Syndrome Components and Breast Density

	FPG		Systolic blood pressure		Diastolic blood pressure		TG		HDL		WC
	R	P	r	P	r	P	r	P	r	P	r	P
BD	−0.13^**	0.005	−0.13^**	0.003	−0.09^*	0.03	0.001	0.978	0.09^*	0.04	−0.09^*	0.02

Bold values: P < 0.05.

Correlation significance for P < 0.05.

Correlation significance for P < 0.01.

Table 4.

Multivariate Logistic Regression Analysis of the Parameters that Could Affect Breast Density

	B coefficient	SE	P	95% CI
	B coefficient	SE	P	Lower	Upper
Age (years)	−0.04	0.03	0.08	0.091	1.01
Gravidity	0.08	0.09	0.31	0.92	1.28
Parity	−0.56	0.16	0.001	0.41	0.78
WC (cm)	0.01	0.01	0.43	0.98	1.03
FPG (mg/dL)	−0.02	0.01	0.12	0.96	1.01
BMI (kg/m²)	−0.11	0.04	0.002	0.83	0.95
TG (mg/dL)	1.40	0.89	0.12	0.71	23.19
HDL (mg/dL)	0.38	1.70	0.82	0.05	41.49

Bold values: P < 0.05.

CI, confidence interval; SE, standard error.

Discussion

There is still lack of evidence to show the relationship between MetS and BD. Up to now, only a few studies have shown different results that seem to differ considering ethnicity.^1,3,10

A study by Kim et al. evaluating the relationship between the presence of MetS and BD in pre and postmenopausal women, nonsignificant relation was observed between MetS and BD in postmenopausal women.³ In another study that comprised women from two different regions of Mexico, MetS was not related to higher BD among postmenopausal women; however, BD was higher among premenopausal women with MetS than in those without MetS.¹ In a multi-ethnicity United States-Study of Women's Health Across the Nation (U.S.-SWAN) study that comprised premenopausal and perimenopausal women, there was no relation between MetS and higher BD scores after adjustment for BMI.¹⁰ Another key point to consider is the likely relation between individual MetS parameters and BD. Most studies in the literature have focused on abdominal adiposity that is a well-known hazardous factor for breast cancer etiopathogenesis.^1,10,11 However, in a study by Vachon et al., which evaluated 1900 mammograms, abdominal adiposity was associated with lower BD results.¹² Moreover, according to the SWAN study, a negative correlation was observed between abdominal fat tissue and BD after adjustment for BMI.¹⁰ In our study, we observed a negative association between WC and BD; besides, lower BMI was significantly related to higher BD scores after adjustments in multivariate regression analysis. Furthermore, significantly higher BD rates were observed in the without MetS group than the with MetS group. However, no significant relation was observed between TG and BD and a negative correlation was observed between WC and BD. With regard to our results, we may speculate that our study population might differ from the previous studies owing to ethnicity, Mediterranean dietary habits, and also the postmenopausal hypoestrogenic state that results in breast atrophy. Moreover, increased BD in the non-MetS group also confirm the hypothesis that obese women could have a nondense image (fatty tissue) compared with dense breast image (parenchymal tissue) at mammography.

HDL and TG are other parameters of MetS and their relation with BD, which is a potential risk factor for breast cancer, have also been investigated owing to the knowledge that increased TG or decreased HDL could be associated with 60% of all postmenopausal breast cancers.^1,13 In addition, higher HDL levels were associated with a decreased risk of breast cancer in a study conducted among Danish postmenopausal women.¹⁴ A Mexican study showed that lower HDL was related to higher BD in one region, whereas there was no relation between HDL or TG and BD among either premenopausal or postmenopausal women in the other region.¹ In another study, there was minimal positive correlation with HDL and BD^15,16 and a modest opposite relation was observed with TG and BD.¹⁶ In our study, similar to the literature mentioned previously, we also observed a positive correlation between HDL and BD; however, no significant correlation was observed for TG and BD scores. According to these results, HDL that transports peripheral cholesterol from bloodstream to liver and reduces white adipose tissue¹⁷ might lead to an increased BD because of lower peripheral tissue adiposity.

Previous reports have mentioned that hyperinsulinemia, glucose consumption, and hypertension could increase the inflammatory state and oxidative stress¹⁸ that lead to breast carcinogenesis in patients with MetS.^19
–21 In a previous study, high blood glucose levels and presence of insulin resistance were positively correlated with BD in both pre and postmenopausal women.¹⁰ In another study, there was no correlation between individual MetS components and BD, but FPG was positively associated with BD.³ However, in the SWAN study, increased glucose levels were associated with lower BD in patients with MetS.¹⁰ In addition, a study conducted among American women revealed that diabetes was associated with lower BD patterns.²² In our study, there were negative correlations between FPG, systolic or diastolic blood pressures and BD. Furthermore, after adjustment for multivariate regression analysis, we observed no significant effect of FPG and blood pressure results on BD. According to our results and the other studies,^3,10,22 we suggest that although increased glucose or insulin resistance could be associated with increased breast cancer risk because of increased growth factors and mitogenesis in the long run,²⁰ their effects may not cause an immediate increase in BD.

The relationship between BD and age was reported by Checka et al. They showed that 74% of the study population between 40 and 49 years of age had an increased BD, whereas this rate decreased to 57% for women in their 50s.²³ As visceral adiposity might seem to be affected by estrogen levels, Sprague et al. reported that there was no relationship between BD and estradiol levels.²⁴ BD may have hormonal etiology at some point such as progesterone and sex hormone binding globulin; however, it may differ from the ethiopathogenesis of breast cancer risk. Our study confirms with previous studies that non-MetS group comprises younger postmenopausal patients with higher BD rates than MetS group.

As a limitation of our study, we did not evaluate the distribution of lifestyle, dietary habit, alcohol consumption, bra size, and socioeconomic state of the study population. To generalize the results of our country, a nationwide study with adequate sample size should be designed considering all variables that may have an effect on BD.

To sum up, the relation between MetS and BD has not been clearly demonstrated in Turkish women. Considering our findings, at least for our region, we suggest that lower BMI and parity may be associated with higher BD in postmenopausal women. In addition, higher HDL and lower FPG, blood pressure, TG, and WC may be correlated with higher BD scores.

Footnotes

Author Disclosure Statement

No conflicting financial interests exist.

References

Rice

, Biessy

, Lajous

, et al. Metabolic syndrome and mammographic density in Mexican women. Cancer Prev Res, 2013; 6:701–710.

Sendur

MAN

, Güven

. Current overview of metabolic syndrome. İç Hastalıkları Dergisi, 2011; 18:125–131.

Kim

, Chang

, Ahn

, et al. Metabolic syndrome, insulin resistance, and mammographic density in pre- and postmenopausal women. Breast Cancer Res Treat, 2015; 153:425–434.

Xue

, Michels

. Diabetes, metabolic syndrome, and breast cancer: A review of the current evidence. Am J Clin Nutr, 2007; 86:823–835.

Boyd

, Lockwood

, Byng

, et al. Mammographic densities and breast cancer risk. Cancer Epidemiol Biomarkers Prev, 1998; 7:1133–1144.

Bjørge

, Lukanova

, Jonsson

, et al. Metabolic syndrome and breast cancer in the me-can (metabolic syndrome and cancer) project. Cancer Epidemiol Biomarkers Prev, 2010; 19:1737–1745.

Rosato

, Bosetti

, Talamini

, et al. Metabolic syndrome and the risk of breast cancer in postmenopausal women. Ann Oncol, 2011; 22:2687–2692.

Expert Panel on

Detection

, Evaluation, and Treatment of High Blood Cholesterol in

Adults

. Executive summary of the third report of the National Cholesterol Education Program (NCEP) expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (Adult Treatment Panel III) final report. JAMA, 2001; 285:2486–2497.

Sickles

, D'Orsi

, Bassett

, et al. ACR BI-RADS^® Atlas, Breast Imaging Reporting and Data System. Reston, VA: American College of Radiology; 2013. ACR BI-RADS® Mammography.

10.

Conroy

, Butler

, Harvey

, et al. Metabolic syndrome and mammographic density: The Study of Women's Health Across the Nation. Int J Cancer, 2011; 129:1699–1707.

11.

Sung

, Song

, Stone

, et al. Association of body size measurements and mammographic density in Korean women: The Healthy Twin study. Cancer Epidemiol Biomark Prev, 2010; 19:1523–1531.

12.

Vachon

, Kuni

, Anderson

, et al. Association of mammographically defined percent breast density with epidemiologic risk factors for breast cancer (United States). Cancer Causes Control, 2000; 11:653–662.

13.

Agnoli

, Berrino

, Abagnato

, et al. Metabolic syndrome and postmenopausal breast cancer in the ORDET cohort: A nested case-control study. Nutr Metab Cardiovasc Dis, 2010; 20:41–48.

14.

Furberg

, Veierod

, Wilsgaard

, et al. Serum high-density lipoprotein cholesterol, metabolic profile, and breast cancer risk. J Natl Cancer Inst, 2004; 96:1152–1160.

15.

Furberg

, Jasienska

, Bjurstam

, et al. Metabolic and hormonal profiles: HDL cholesterol as a plausible biomarker of breast cancer risk. The Norwegian EBBA Study. Cancer Epidemiol Biomarkers Prev, 2005; 14:33–40.

16.

Boyd

, Connelly

, Byng

, et al. Plasma lipids, lipoproteins, and mammographic densities. Cancer Epidemiol Biomarkers Prev, 1995; 4:727–733.

17.

McGillicuddy

, Reilly

, Rader

. Adipose modulation of high-density lipoprotein cholesterol implications for obesity, high-density lipoprotein metabolism, and cardiovascular disease. Circulation, 2011; 124:1602–1605.

18.

Cowey

, Hardy

. The metabolic syndrome: A high-risk state for cancer?. Am J Pathol, 2006; 169:1505–1522.

19.

Abdilla

, Tormo

, Fabia

, et al. Impact of the components of metabolic syndrome on oxidative stress and enzymatic antioxidant activity in essential hypertension. J Hum Hypertens, 2007; 21:68–75.

20.

Menon

, Ram

, Dorn

, et al. Oxidative stress and glucose levels in a population based sample. Diabet Med, 2004; 21:1346–1352.

21.

Muti

, Quattrin

, Grant

, et al. Fasting glucose is a risk factor for breast cancer: A prospective study. Cancer Epidemiol Biomarkers Prev, 2002; 11:1361–1368.

22.

Roubidoux

, Kaur

, Griffith

, et al. Correlates of mammogram density in southwestern Native-American women. Cancer Epidemiol Biomarkers Prev, 2003; 12:552–558.

23.

Checka

, Chun

, Schnabel

, et al. The relationship of mammographic density and age: Implications for breast cancer screening. AJR Am J Roentgenol, 2012; 198:292–295.

24.

Sprague

, Trentham-Dietz

, Gangnon

, et al. Circulating sex hormones and mammographic breast density among postmenopausal women. Horm Cancer, 2011; 2:62–72.